Description :

AI/ML Engineer

Exp : 8 to 10 years

Location : Pune

Key Responsibilities :

- Design, develop, and optimize AI/ML algorithms for medical image analysis, segmentation, and 3D reconstruction from TEE and CT images.

- Research and implement advanced deep learning architectures including CNNs, GANs, VAEs, and Diffusion Models for medical imaging tasks.

- Develop robust 3D reconstruction pipelines from 2D image data and multi-view geometries, tailored to medical imaging workflows.

- Perform multimodal image registration (CT-CT, CT-MRI, Fluro-Endo, 2D-3D) and develop tools for alignment, calibration, and fusion.

- Enhance and denoise medical images using advanced computer vision and AI-based enhancement techniques.

- Work extensively with DICOM data, integrating with PACS systems for data ingestion and retrieval.

- Collaborate with teams for dataset curation, labeling, and ground truth generation.

- Develop scalable training and inference pipelines on cloud platforms (AWS preferred; Azure/GCP acceptable).

- Ensure reproducibility and traceability in experiments using MLOps practices (Docker, MLflow, or similar).

- Collaborate with software engineers to integrate AI components into production-grade imaging applications.

- Document research findings, maintain version-controlled repositories, and contribute to technical publications or IP filings.

- Stay up-to-date with emerging trends in AI, computer vision, and medical imaging technologies.

- Bachelors, Masters, or Ph.D in Computer Science, Data Science, Biomedical Engineering, or related field with focus on AI, ML, or Computer Vision.

- 8+ years of hands-on experience in AI/ML model development with strong exposure to computer vision and imaging applications.

- Expert-level proficiency in Python and deep learning frameworks such as PyTorch and TensorFlow.

- Experience in 2D and 3D medical imaging (CT, MRI, Ultrasound, TEE) and DICOM data handling.

- Strong understanding of 3D geometry, camera calibration, stereo vision, and multi-view reconstruction.

- Experience in segmentation, registration, and object tracking within medical image contexts.

- Proficiency with classical computer vision techniques (OpenCV, PCL, feature detection, structure-from-motion, SLAM, etc.

- Knowledge of generative and reconstruction models (GANs, VAEs, Diffusion Models) and fine-tuning methods for domain-specific applications.

- Experience with data preprocessing, augmentation, and pipeline automation for large-scale medical datasets.

- Familiarity with MLOps, containerization (Docker), and deployment workflows for cloud and edge environments